Assessing whether residence-based static exposure can represent activity-based exposure is important for refined air pollution exposure assessment, particularly for vulnerable populations such as primary schoolchildren. We quantified static–dynamic exposure differences among primary schoolchildren from 22 primary schools and 138 residential communities and examined their associations with housing price in Luozhuang District, Linyi, China. To support exposure estimation, we developed high-resolution land-use regression (LUR) models integrating dense monitoring observations, road networks, area-of-interest data, and other spatial predictors to generate 100 m × 100 m typical hourly concentrations of PM2.5, PM10, NO2, and O3. After incorporating children’s school-time locations, dynamic exposure estimates remained broadly comparable to static estimates but showed systematic pollutant-specific differences. Significant differences were found for NO2, PM2.5, and O3 in both seasons, with small absolute mean differences of 0.04–0.19 μg/m3. Static estimates tended to overestimate exposure to NO2 and O3 but underestimate exposure to PM2.5. Housing-price-stratified analyses suggested relatively smaller static–dynamic exposure differences in low-price communities and a positive housing-price trend for warm-season O3, although no robust linear association was observed overall. These findings suggest that residence-based exposure estimates can broadly approximate primary schoolchildren’s daily exposure, although ignoring school-time locations may introduce small pollutant-specific deviations in refined exposure assessment.